Cloudera demystifies data management in the age of AI

Abhas Ricky, chief strategy officer at leading hybrid data company Cloudera, shares how companies are leveraging enterprise AI to unlock richer, more accurate, and more secure data insights.

In the race to integrate AI and machine learning (ML) models into any given enterprise, leaders are prioritizing data management. Gathering and storing clean, structured data has become a primary objective to train AI/ML models and make better decisions. Gartner predicts that 80% of organizations will need to employ multiple data hubs by 2026 to serve their data and analytics needs, an expensive and technically complex endeavor for even the most advanced organization. This is where Cloudera, a leading hybrid cloud data management company, is paving the way.

Founded in 2008 as an open-source data storage startup, Cloudera has evolved into a popular hybrid cloud solution for top companies across various industries, from finance to health care. Today, the company, which specializes in providing scalable and secure data management platforms with portable, cloud-native data services, has been at the forefront of helping businesses harness their data for AI-driven insights.

Fortune Brand Studio recently sat down with Abhas Ricky, chief strategy officer at Cloudera, to discuss the unique data management challenges enterprises face in adopting AI and how Cloudera is poised to meet these evolving needs. With more than eight years at the company, Ricky offers a unique perspective on the opportunities and pitfalls enterprise leaders should be monitoring and shares how they can take charge of their data infrastructure to keep pace in the AI race.

What are some of the biggest challenges enterprises are facing in adopting AI?

There are three core areas that large enterprises are struggling with. One is accessing the high-fidelity data they need to train their AI models. The higher the fidelity of the data, the higher the fidelity of the outputs that you’re going to get. However, getting access to that clean data is not always easy. The second one is skillsets. Most people are still new to the enterprise AI game. It’s hard for large enterprises to hire people who know how to deal with the sudden burst of capabilities that generative AI (gen AI) and machine learning (ML) tools bring. And third is the high cost of computing and testing. There are companies that are saving millions of dollars of compute cost a month just by running their workloads on GPUs and private cloud. 

You’ve been with Cloudera for more than eight years now, which means you’ve witnessed several stages of the AI evolution. How has Cloudera evolved along with these trends to meet the needs of enterprise customers?

With the evolution of AI, a lot has changed in terms of scale at Cloudera. We’ve evolved from data storage to analytics and data management to open-source support. Fast-forward to today, we’re now one of the world’s largest enterprise AI companies with more than 25 exabytes of data, serving mission critical use cases worldwide—including banking and telecommunications computing needs. These stages have required us to make big leaps in terms of how our engineers build products and how our end practitioners and customers use our products.

How does Cloudera’s hybrid approach help organizations achieve their AI goals?

A majority of large organizations do analytical workloads in a public cloud and operational workloads in a private cloud. But oftentimes, they will start with public cloud and bring it back to private cloud to streamline costs and for security and governance benefits. Application portability—the ability to move applications between public and private clouds without the need to refactor them—is what allows us to offer true hybrid capabilities. 

On another note, a hybrid cloud strategy is not just for the analytics. It’s also for AI. In the AI world, there’s an even bigger focus on data because that’s what drives the better output. One of our mantras at Cloudera is, “You have to bring the models to the data and not the data to the models.” That’s the capability we provide.

Enterprise businesses are prioritizing security, compliance, and accuracy in their AI strategies. How is Cloudera supporting or facilitating these features in its offerings?

Cloudera has focused on these two strengths for years: having the best total cost of ownership (TCO), and the best security and metadata governance. For example, our shared data experience  product is a unified common data plan that allows a 360-degree view of your data assets around lineage, provenance, authorization, and even role-based access control. You can tell which data was created by whom, where it was copied, who should have access to it, and so on. Those things are critical for security and privacy. Add to that our capabilities around metadata governance, both technical and business metadata, and you have a full suite of governance solutions. That function is fundamental to what we do and a core part of the Cloudera.. That’s one of the reasons why customers buy us.

What are some of the most impactful results you’ve seen that the Cloudera platform has facilitated with enterprises?

One of the most strategic customers we have in the enterprise AI arena is the Oversea-Chinese Banking Corporation (OCBC), the second largest financial services group in Southeast Asia by assets. It’s been adapting large language models (LLMs) to its needs and running them on the Cloudera Machine Learning platform using NVIDIA GPUs. When the ChatGPT phenomenon started, OCBC was ahead of the game in using AI for text summarization, chatbots, and AI agents. It used to take around 25 minutes for relationship managers to conduct research to generate investment talking points for customers–now it takes two minutes. Its chatbot, Chat Q&A, has been supporting customer call center operations and its copilots have helped improve engineering productivity for some time.

Cloudera announced several new partnerships at the EVOLVE24 New York event in October. Why do you think industry collaboration is so important in the AI space right now?

AI is a team sport. There is no one company that can provide all capabilities and solutions thoroughly. Through our partnerships with companies such as NVIDIA, Google Cloud, Anthropic, Snowflake, and AWS, Pinecone, and others, we are helping customers build capabilities across the life cycle of their AI applications.

There are the five stages to building AI applications and agentic applications: natural language processing (NLP) capability, model deployment, vectorization, the core platform to serve the data, and prompting (or fine-tuning) capability. When we work with partners, we’re able to integrate well and serve our joint customers with the support from a group of expert advisors across all these companies, not just us. It enables faster enterprise AI application deployment at a better price.

What is Cloudera’s long-term vision for AI and data management? How will the company continue to lead the sector?

We want to make sure that customers can build the latest generation of AI applications and agentic applications by leveraging the proprietary data that they have. And we will continue building the tooling they need.

Additionally, we do think that the world is moving toward “cognition-as-a-service.” Businesses need more than just data; they need insights and value that drives actions. So we will add capabilities to offer not just data management but also enable knowledge and insights. That’s where we think AI agents will really shape the next wave of AI. We want to provide those cognitive frameworks and capabilities because that’s where the value is.